Low-complexity detection and performance in multi-gigabit high spectral efficiency wireless systems

Abstract

Recently, we reported a low-complexity likelihood ascent search (LAS) detection algorithm for large MIMO systems with several tens of antennas that can achieve high spectral efficiencies of the order of tens to hundreds of bps/Hz. Through simulations, we showed that this algorithm achieves increasingly near SISO AWGN performance for increasing number of antennas in i.i.d. Rayleigh fading. However, no bit error performance analysis of the algorithm was reported. In this paper, we extend our work on this low-complexity large MIMO detector in two directions: i) We report an asymptotic bit error probability analysis of the LAS algorithm in the large system limit, where N_t, N_r → ∞ keeping N_t = N_r, where N_t and N_r are the number of transmit and receive antennas, respectively. Specifically, we prove that the error performance of the LAS detector for V-BLAST with 4-QAM in i.i.d. Rayleigh fading converges to that of the maximum-likelihood (ML) detector as N_t, N_r → ∞ keeping N_t = N_r. ii) We present simulated BER and nearness to capacity results for V-BLAST as well as high-rate non-orthogonal STBC from Division Algebras (DA), in a more realistic spatially correlated MIMO channel model. Our simulation results show that a) at an uncoded BER of 10⁻³, the performance of the LAS detector in decoding 16×16 STBC from DA with N_t = N_r = 16 and 16-QAM degrades in spatially correlated fading by about 7 dB compared to that in i.i.d. fading, and b) with a rate-3/4 outer turbo code and 48 bps/Hz spectral efficiency, the performance degrades by about 6 dB at a coded BER of 10⁻⁴. Our results further show that providing asymmetry in number of antennas such that N_r » N_t keeping the total receiver array length same as that for N_r = N_t, the detector is able to pick up the extra receive diversity thereby significantly improving the BER performance.